This invention is in the field of plant molecular biology. More specifically, this invention pertains to the modification of starch biosynthetic gene expression to produce starches in plants and seeds.
Starch is a mixture of two polysaccharides, amylose and amylopectin. Amylose is an unbranched chain of up to several thousand xcex1-D-glucopyranose units linked by xcex1-1,4 glycosidic bonds. Amylopectin is a highly branched molecule made up of up to 50,000 xcex1-D-glucopyranose residues linked by xcex1-1,4 and xcex1-1,6 glycosidic bonds. Approximately 5% of the glycosidic linkages in amylopectin are xcex1-1,6 bonds, which leads to the branched structure of the polymer.
Amylose and amylopectin molecules are organized into granules that are stored in plastids. The starch granules produced by most plants are 15-30% amylose and 70-85% amylopectin. The ratio of amylose to amylopectin and the degree of branching of amylopectin affects the physical and functional properties of the starch. Functional properties, such as viscosity and stability of a gelatinized starch determine the usefulness and hence the value of starches in food and industrial applications. Where a specific functional property is needed, starches obtained from various crops such as corn, rice, potatoes or wheat may meet the functionality requirements. If a starch does not meet a required functional property, such as the need for stable viscosity under high temperatures and acidic conditions, the functionality can usually be achieved by chemically modifying the starch. Various types and degrees of chemical modification are used in the starch industry, and the labeling and use of chemically modified starches must meet government regulations.
Within the starch bearing organs of plants, the proportion of amylose to amylopectin and the degree of branching of amylopectin are under genetic control. For example, corn plants homozygous for the recessive waxy (wx) mutation lack a granule-bound starch synthase enzyme and produce nearly 100% amylcpectin. Corn plants homozygous for the recessive amylose extender (ae) mutation and uncharacterized modifier genes can reportedly produce starch granules that are approximately 80% to 90% amylose (see U.S. Pat. No. 5,300,145). The dull mutant of corn lacks a starch synthase distinct from that lacking in the waxy lines and has a starch characterized by more amylose and a larger proportion of shorter branches on the amylopectin molecule than normal starch.
Most cereal crops are handled as commnodities, and many of the industrial and animal feed requirements for these crops can be met by common varieties which are widely grown and produced in volume. However, there exists at present a growing market for crops with special end-use properties which are not met by grain of standard composition. Most commonly, specialty corn is differentiated from xe2x80x9cnormalxe2x80x9d corn by altered endosperm properties, such as an overall change in the ratio of amylose to amylopectin as in waxy or high amylose corn, an increased accumulation of sugars as in sweet corn, or an alteration in the degree of endosperm hardness as in food grade corn or popcorn (Glover, D. V. and E. T. Mertz (1987) in Corn: Nutritional Quality of Cereal Grains; Genetic and Agronomic Improvement, R. A. Olson and K. J. Frey, eds. American Society of Agronomy, Madison Wis., pp. 183-336; Rooney, L. W. and S. O. Serna-Saldivar (1987) Food Uses of Whole Corn and Dry-milled Fractions, in Corn: Chemistry and Technology, S. A. Watson and P. E. Ramstead, eds. American Association of Cereal Chemists, Inc., St. Paul, Minn., pp. 399-429). The current invention offers the buyers of specialty grains a source of starch having properties distinct from waxy starch and offers farmers the opportunity to grow a higher value-added crop than normal or waxy corn.
Purified starch is obtained from plants by a milling process. Corn starch is extracted from kernels through the use of a wet milling process. Wet milling is a multi-step process involving steeping and grinding of the kernels and separation of the starch, protein, oil and fiber fractions. A review of the corn wet milling process is given by S. R. Eckhoff (1992) in the Proceedings of the Fourth Corn Utilization Conference, June 24-26, St. Louis, Mo., printed by the National Corn Growers Association, CIBA-GEIGY Seed Division and the United States Department of Agriculture. Wheat is also an important source of purified starch. Wheat starch production is reviewed by J. W. Knight and R. M. (1984) Olson in Starch: Chemistry and Technology 2nd Edition., Academic Press. Eds. Whisler et al.
Starch is used in numerous food and industrial applications and is the major source of carbohydrates in the human diet. Typically, starch is mixed with water and cooked to form a thickened gel. This process is termed gelatinization. Three important properties of a starch are the temperature at which gelatinization occurs, the viscosity the gel reaches, and the stability of the gel viscosity over time. The physical properties of unmodified starch during heating and cooling limit its usefulness in many applications. As a result, considerable effort and cost is needed to chemically modify starch in order to overcome these limitations of starch and to expand the usefulness of starch in industrial applications.
Some limitations of unmodified starches and properties of modified starches are given in Modified Starches: Properties and Uses, O. B. Wurzburg, ed., (1986) CRC Press Inc., Boca Raton, Fla. Unmodified starches have very limited use in food products because the granules swell and rupture easily, thus forming weak bodied, undesirable gels. Chemical modifications are used to stabilize starch granules thereby making the starch suitable for thousands of food and industrial applications including baby foods, powdered coffee creamer, surgical dusting powders, paper and yarn sizings and adhesives. Common chemical modifications include cross linking, in which chemical bonds are introduced to act as stabilizing badges between starch molecules, and substitution in which substituent groups such as hydroxyethyl, hydroxypropyl or acetyl groups are introduced into starch molecules.
The use of chemically modified starches in the United States is regulated by the Food and Drug Administration (FDA). xe2x80x9cFood starch-modifiedxe2x80x9d starches may be used in food but must meet specified treatment limits, and xe2x80x9cindustrial starch-modifiedxe2x80x9d starches may be used in items such as containers that come in contact with food and must also meet specified regulatory requirements; Code of Federal Regulations, Title 21, Chapter 1, Part 172, Food Additives Permitted in Food for Human Consumption, Section 172, 892, Food Starch-Modified, U.S. Government Printing Office, Washington, D.C. 1981; (a) Part 178, Indirect Food Additives, Sect. 178.3520, Industrial Starch-Modified. These regulations limit the degree of chemical modification by defining the maximum amount of chemical reagent that can be used in the modification steps. The levels of by-products in starch resulting from the modification process are also regulated. For example, propylene chlorohydrin residues in hydroxypropyl starch are of special concern (Tuschhoff, J. V. (1986) Hydroxypropylated Starches, in Modified Starches: Properties and Uses, O. B. Wurzburg, ed., CRC Press, Boca Raton, Fla., pp. 55-57).
In addition to its use as a purified ingredient, starch is an important component of whole flours, such as wheat flour, used in the production of breads, baked goods and pastas. Starch comprises between 50 and 70% of the weight of a wheat grain and its importance in the performance of wheat flours is well known in the art. Although the complex genetics of wheat has limited the variations in starch fine structure that is available in whole flours, the production of novel starch structures in wheat or other flours may result in improved performance of these whole flours in food product applications. Starch structure is also an important component of the quality of whole consumed cereal grains such as rice. Differences in amylopectin fine structure have been related to cooked rice texture (Reddy et al. (1993) Carbohydr. Polymers 22:267-275).
Differences in the degree of starch branching or polymerization are known to result in a change in the physiochemical properties of starch. It has been suggested that starches, tailor-made for specific applications, may be generated by alteration of the branch chain distribution of the amylopectin molecule, the relative proportion of amylose to amylopectin or the degree of polymerization of anylose. Some authors (Shi and Seib (1992) Carbohydr. Res. 227:131-145; Jane et al. (1999) Cereal Chemistry In Press), have reported that. retrogradation tendency is reduced in starches from different botanical sources which contain increased proportions of very short chains (DP 6-9) in their amylopectin, but no suggestion as to how this might be achieved in corn is made. However, achieving phenotypic alteration of starch composition has been problematic; while key enzymes in starch biosynthesis have been identified, their exact roles remain uncertain. Thus, correlation of activities of particular enzymes with particular molecular characteristics of starch structure and, in turn, with starch function in food and industrial products has been difficult. Although desirable functional properties that an ideal starch might need can be envisioned, there is only a limited understanding of what the molecular structure of the starch should be to achieve this and little understanding of how particular starch biosynthetic enzymes specifically affect those parameters. For example, the role of individual enzymes in determining the branching patterns and length of branches is as yet unclear and is compounded by the lack of understanding of how branching enzymes and starch synthases interact. In addition, while the role of the granule-bound starch synthase encoded by the waxy gene is fairly well understood; see Denyer et al. (1996) Plant J. 10:1135-1143), the number and exact functions of other starch synthases, soluble or granule-bound, are not well understood. (Smith et al. (1996) Ann. Rev. Plant Phys. and Mol. Biol. 48:67-87).
WO 94/09144 discusses the generation of corn plants with improved ability to synthesize starch at elevated temperatures. This publication proposes that the limiting factor in grain filling at elevated temperature is the lability of certain starch biosynthetic enzymes, particularly starch synthase (SS) and starch branching enzyme (SBE). The introduction of genes encoding enzymes that have a higher optimum temperature for activity or that have a higher tolerance to heating into plants may afford an increase in the amount of starch deposited in the corn kernel. Moreover, it is claimed that this strategy may be used to generate starch of altered fine structure as a result of the introduction of donor genes whose expression may alter the balance of the different starch biosynthetic enzymes. Suggested donor genes include those that encode enzymes that display improved kinetic or allosteric properties relative to the endogenous enzyme or an extra copy of the endogenous gene that would compensate for losses in enzyme activity incurred due to heat lability. As a means to alter starch structure, WO 94/09144 also suggests the use of sense and antisense genes to alter the natural ratios of the different starch synthase and branching enzymes in the recipient plant. This publication discloses the effect of temperature on catalytic activity and enzyme stability for certain starch biosynthetic enzymies. However, no data are presented to substantiate the proposed molecular strategies. Indeed, while this publication suggests the use of altered starch synthase expression to alter starch fine structure, both amylose/amylopectin ratios and degree of amylopectin branching, other publications before and after suggest that starch branching enzymes, not just starch synthases, would be required, or that still other factors must be addressed. For example, Smith et al. (1995, Plant Phys. 107:673-677) suggest two distinct views about the determination of the branching pattern of amylopectin: first, that the pattern represents a balance between the activities of branching and debranching enzymes, and second, that the pattern can be explained largely by the properties of branching enzymes. No role for starch synthases is provided. Guan and Preiss (1993, Plant Phys. 102:1269-1271) suggest a study of the interactions among the multiple forms of branching enzymes and starch synthase may be essential for understanding the specificity and function of the individual isozymes and the mechanism of amylopectin biosynthesis. Thus, Guan and Preiss imply the need to alter both enzymes at once. Lastly, Van den Koornhuyse et al. (1996, J. Biol. Chem. 271:16281-16287) propose that low nucleotide sugar concentrations are either directly or indirectly responsible for the major differences observed in the composition or structure of starch during storage. In sum, it is clear from the differing views that there is no consensus as to exactly what factors affect starch structure and thus how to alter it. Furthermore, no workers, including WO 94/09144, present evidence demonstrating that soluble starch synthases limit the rate of polymerization and therefore that either increasing or reducing their level will actually alter starch fine structure. WO 94/09144 further does not teach how to differentiate between genes encoding isoforms that make a minimal contribution to starch biosynthesis and more active forms. Reducing the expression pattern of a relatively inactive (at the enzymatic level, not necessarily at the transcriptional level) enzyme is unlikely to have an effect. In sum, WO 94/09144 makes a suggestion but does not teach in sufficient detail for the skilled artisan to actually produce a starch altered in fine structure.
There have been several reports of alteration of starch structure by modification of SBE expression in both potato (Virgin et al. (June, 1994) at the 4th International Congress of Plant Molecular Biology, and Christensen et al. and Kossman et al. (July, 1994) at the Plant Polysaccharide Symposiumn) and corn (Broglie et al.; WO 97/22703). None of that work addresses the potential of altering the expression of starch synthases. Several authors have speculated that altering non-granule-bound starch synthase I (non-GBSSI) synthase expression would alter starch structure or compostion, but this has not been clearly demonstrated in cereals (Block et al., WO 9745545A; Frohberg and Kossmann, WO 9744472; Frohberg and Kossmann, WO 9726362).
Although the enzymatic steps are known, the molecular details of starch biosynthesis are not well understood. It is not clear whether the different SS isoforms contribute equally throughout starch biosynthesis or whether each isoform plays a distinct role in assembling the amylopectin molecule at discrete steps along an obligatory pathway. In consideration of the possible interplay between the starch branching enzymes and the multiple starch synthases that function in glucan chain elongation, it is impossible to make accurate predictions concerning starch structure based upon the catalytic properties of each enzyme.
Beyond the clear role of GBSSI in amnylose biosynthesis, the exact roles of individual starch synthases are not clear. There is evidence from some, but not all, species that individual isoforms of SS make qualitatively different contributions to amylopectin biosynthesis and that GBSSI may also contribute to amylopectin as well as amylose biosynthesis (Smith et al. (1996) Ann. Rev. Plant Phys. and Mol. Biol. 48:67-87). Numerous starch synthases have been cloned from different species, but Edwards et al. (1960, Plant Phys. 112:89-97), and Mu-Forster et al. (1996, Plant Phys. 111:821-829) demonstrated that the previously made distinction between granule-bound and soluble starch synthases may not reflect the in vivo situation and it will not be used here with the exception of the waxy protein, GBSSI. Since its original isolation from corn (Klxc3x6sgen et al. (1986) Mol. Gen. Genet. 203:237-244) this gene has been cloned from many species. Numerous other SS have been cloned from a range of species, but they appear to be less closely related across species than GBSSI. Potato contains at least two other starch synthases, SSII and SSIII (Marshall et al. (1996) Plant Cell 8:1121-1135). Pea contains a synthase designated SSII which appears to be present in two forms, one derived by the processing of the other (Edwards et al. (1995) Plant Phys. 112:89-97). Block et al. (WO 9745545A) isolated two starch synthase cDNA clones from wheat. Three forms of soluble starch synthase were purified from rice. These were shown to be derived from a primary form by the isolation of the corresponding gene referred to as soluble starch synthase I (SSSI) (Baba et al. (1993) Plant Phys. 103:565-573; Tanaka et al. (1995) Plant Phys. 108:677-683). Rice Expressed Sequence Tags (ESTs) showing homology to the starch synthase II sequences of pea and potato have been identified (AA752475, AA753266, AA751702, AA751557, AA751512A; Nahm, B. H. at al.). A sequence related to rice SSI was isolated from corn (Harn et al. (1995) Plant Phys. 108:S-50; Keeling et al., WO 9720936) and was designated corn SSI. Two additional starch synthase cDNA clones have been isolated by Keeling et al. (WO 9720936). The expression of the genes encoding these starch synthases has been studied and their representation in the corn genome has been reported (Harn et al. (1998) Plant Mol. Biol. 37:639-649). Frohberg and Kossmann (WO 97/44472 and WO 97/26362) have also reported the isolation of two of these corn starch synthase sequences. The locus responsible for the dull mutation in corn was recently shown to encode another starch synthase (Gao et al. (1998) Plant Cell 10:399-412). In a study characterizing the soluble starch sythase activities in maize endosperm Cao et al. (1999, Plant Physiol. 120:205-215) report that DUI and SSSI likely account for all of the soluble starch synthase activity in developing endosperm. Unicellular organisms also contain multiple starch synthases (Fontaine et al. (1993) J. Biol. Chem. 268:16223-16230; Buleon et al. (1997) Plant Physiol. 115:949-957). While for some of these enzymes authors have speculated on its particular role, in no case has it been elucidated how the full complement of starch synthase isoforms work together to elongate amylopectin branches or how the entire array of starch biosynthetic enzymes in a particular species interact and function together to produce the starch structure that is observed in the mature seed or tuber. In particular, the role of low abundance starch synthases in endosperm is unclear.
It is well known that the waxy mutation in corn results in the lack of a functional GBSSI enzyme and in altered starch composition. Similarly in wheat, low amylose vareties have been selected which lack the GBSS. Dominant forms of the analogous mutation in potato have been made by expressing GBSSI antisense genes in transgenic potato plants (Visser et al. (1989) Plant Mol. Biol. 17:691-699). Shewmaker et al. (1994, Plant Physiol. 104:1159-1166) have reported altered starch structure in potato through the expression of E. coli glycogen synthase in tubers. Modified expression of non-GBSSI starch synthases have also been reported in potato. Reduction of SSII expression in transgenic potato tubers has been achieved using antisense technology. Decreased levels of SSII protein were not correlated with any detectable change in starch content or composition and starch granule morphology appeared normal (Edwards et al. (1995) Plant J. 8:283-294). Recently, small changes in amylopectin branch chain distribution (dp 6-35) have been reported in SSII antisense potato plants (Edwards et al. (.1999) Plant J. 17:251-261). Different results were observed when the major soluble starch synthase activity of potato tubers, SSIII was inhibited by an antisense approach. In these transgenic plants, starch content and composition was not changed, however, starch granule morphology was noticeably affected (Marshall et al. (1996) Plant Cell 8:1121-1135). Changes in amylopectin branch chain distribution were also observed but these were distinct from those found in the SSII antisense plants (Edwards et al. (1999) Plant J. 17:251-261). A pea mutant, rug5 was found to be lacking in a starch synthase isoform that is highly homologous to the SSII of potato. Although the two starch synthases share homology in amino acid sequence, different results were produced when this starch synthase isoform was inhibited in pea. Noticeable changes were apparent in short (dp  less than 15), medium (dp 15-45) and very long (dp xcx9c1000) amylopectin branch chains. In addition, these structural changes were associated with gross changes in starch granule morphology. Thus, while transgenic results from potato suggest that within a specific organ, different isoforms of starch synthase perform different roles in starch biosynthesis, results obtained from the pea rug5 mutant indicate that homologous isoforms may not necessarily perform the same function in different starch storing organs. Generalization about the role of specific isoforms and prediction of starch phenotypic changes which accompany modified expression is rendered difficult due to differences in the number of isoforms represented within different organs as well as differences in the relative amounts of activity contnrbuted by the different isoforms. While transgenic work aimed at the modification of starch synthase expression has been reported in potato, no similar experiments have been described in corn.
U.S. Pat. No. 5,824,790 reports the isolation and sequence of 3 non-waxy starch synthase cDNA clones from maize. It suggests the use of these sequences to generate constructs designed to modify expression of these starch synthases in transgenic plants. It further suggests that modified protein expression may give rise to a change in the fine structure of the starch. While the nucleotide and protein sequences for the 3 starch synthases are provided, no data are given concerning the generation and characterization of transgenic plants carrying DNA constructs derived from these sequences; similarly no data relating to starch composition and structure in transgenic plants is reported. In the absence of specific roles for the different isoforms of soluble starch synthase in cereal endosperm, and with the prescence of activities for the SSIIa and SSIIb class enzymes in endosperm in question, (Cao et al. (1999) Plant Physiol. 120:205-215) it is clear that gene sequences alone provide no clear indication of what type of change, if any, to starch structure may be accomplished by altering the expression of a particular soluble starch synthase gene. And in the abscense of this predictive power or the actual production of the starch the utility of any given change is unclear. In fact, in terms of specific functional attributes such as retrogradation tendency it is clear that some starch structural changes are actually detrimental to utility. Qiange and Thompson (1998, Carbohydr. Res. 314:221-235) examined retrogradation of three double mutants of maize, duwx, aewx and su2wx, in comparison to normal waxy starch and showed increased retrogradation tendency in two of the three amylopectin types. Thus, it is clear that change alone is insufficient to improve the utility of cereal straches, and that some changes may be improvements while others are neutral or even detrimental. In the absence, of the ability to meaningfully predict the structural change that can be produced with a given genetic modification the only way to identify useful changes is to actually produce the modified starch.
Molecular genetic solutions to the generation of starches from cereal crops with altered fine structures have a decided advantage over more traditional plant breeding approaches. Changes to starch fine structure can be produced by specifically inhibiting expression of one or more of the SS or SBE isoforms by antisense inhibition or cosuppression (WO 94/09144). An antisense or cosuppression construct would act as a dominant negative regulator of gene activity. While conventional mutations can yield negative regulation of gene activity these effects are most likely recessive. The dominant negative regulation available with a transgenic approach may be advantageous for certain grain production methods. In addition, the ability to restrict the expression of the altered starch phenotype to the reproductive tissues of the plant by the use of specific promoters may confer agronomic advantages relative to conventional mutations which will have an effect in all tissues in which the mutant gene is ordinarily expressed. Finally, the variable levels of antisense inhibition or cosuppression that arise from chromosomal position effects could produce a wider range of starch phenotypes than those that result from dosage effects of a mutant allele in cereal endosperm.
The incomplete understanding of the role of different starch synthase enzymes in cereal crops render attempts to manipulate starch fine structure by inhibition of starch synthase gene expression difficult. However, manipulation of starch synthase enzyme gene expression by cosuppression and antisense technology is possible and would likely produce a desirable phenotype. Thus, one of ordinary skill in the art only has to screen multiple transgenic plants for the desired alteration in starch fine structure.
The instant invention discloses utilization of cDNA clones to construct chimeric sense and antisense genes for alteration of starch synthase enzymatic activity in corn grain or endosperm and the grain or endosperm of other cereal crops. More specifically, this invention concerns a method of producing a transformed cereal crop wherein the starch fine structure derived from a grain of the cereal crop is altered compared to the fine structure of starch derived from a non-transformed cereal crop comprising: (1) preparing a chimeric gene comprising a nucleic acid fragment encoding a non-GBSSI starch synthase enzyme structural gene or a fragment thereof, operably linked in either sense or antisense orientation on the upstream side to a nucleic acid fragment encoding a promoter that directs gene expression in endosperm tissue, and operably linked on the downstream side to a nucleic acid fragment encoding a suitable regulatory sequence for transcriptional termination, and (2) transforming cereal crops with said chimeric gene, wherein expression of the chimeric gene results in alteration of the fine structure of starch derived from the grain of the transformed cereal crops compared to the fine structure of starch derived from cereal crops not possessing said chimeric gene. The invention also concerns a method of producing a transformed cereal crop wherein the starch fine structure derived from a grain of the cereal crop has a change in the relative proportions of amylose to amylopectin relative to that of starch derived from cereal crops not possessing the chimeric gene above, or a change in the degree of polymerization of the amylose component of starch derived from the transformed cereal crop relative to the degree of polymerization of the amylose of starch derived from cereal crops not possessing the chimeric gene above. To date no reports have demonstrated an alteration in molecular structure of starch created by altering the expression level of non-GBSSI starch synthase in a transgenic plant. This invention describes the specific alterations in starch structures, changes in amylose to amylopectin ratio, changes in amylopectin fine structure, increased abundance of very short amylopectin chains (DP 6-9), and change in the degree of polymerization of amylose, that can be created by controlling the expression of non-GBSSI starch synthases in transgenic plants.
This invention also concerns cereal crop varieties prepared by transformation using said method, starch isolated from the grain of a cereal crop variety prepared using the above method, and a method of preparing a thickened foodstuff comprising combining a foodstuff, water, and an effective amount of a starch isolated from the grain of a cereal crop variety prepared using the method, and cooking the resulting composition as necessary to produce a thickened foodstuff.
This invention also concerns cereal crop varieties prepared by transformation using the above method, flours prepared from the grain of said cereal crop varieties, and the preparation of breads, baked goods, and pastas by combining water, food ingredients, and an effective amount of flour from the grain of cereals crop varieties prepared using the method, and cooking the resulting composition as necessary to produce a bread, baked good, or pasta product.